281 CERAMIDE-NP IN SKIN-SIMULATING LIPOSOME FORMULATIONS
proportionality tests were evaluated based on the residual variance using the student t-test
(p =0.05). The residues were calculated based on the difference between theoretical and
experimental values, which were estimated from the calibration curve.18
LOD and LOQ. The LOD and LOQ values of the method were calculated from the
calibration curves according to the following equations based on the standart deviation of
the response and slope:17
LOD SD
S =× 3 3 .
LOQ SD
S =× 10
SD is the mean standard deviation of y-axis interception values of calibration curves
S is the mean angular coefficient of calibration curves.
Precision. The precision of the method was evaluated for two levels: repeatability (intra-
assay) and intermediate (inter-assay) precision. The intra-assay precision was performed
with liposome analytes at 80%, 100%, and 120% (320, 400, 480 µg/mL) using three
replicates within a day which qualified the ICH Q2(R1) specifications.16 The inter-assay
precision was performed on three different days at analyte concentrations of (320, 400, and
480 µg/mL) in three replicates using refresh samples prepared by the same analysis. The
results of precision were calculated as the coefficient of variation (CV) in each level for each
analyte concentration, using the following equation:
CV deviation of the peak areas for each analyte/ =(standard
a average of the peak areas for each analyte) × 100
Accuracy. The accuracy of the method was carried out at 80%, 100%, and 120% of
liposome analytes (320, 400, 480 µg/mL) of Cer-NP based on ICH Q2(R1).16 The liposome
formulation including a 600 µg/mL concentration of analytes was prepared with the
appropriate dilutions. The experiments were performed in three replicates. The results of
accuracy were calculated based on the peak areas of Cer-NP in terms of recovery (R), as
described in the following equation:
R measured concentration in the liposomes/
measured concent
=(
tration in the solutions) × 100
RESULTS AND DISCUSSION
METHOD DEVELOPMENT AND OPTIMIZATION OF CHROMATOGRAPHIC CONDITIONS
Three compounds (phosphatidylcholine, Cer-NP, and cholesterol) exhibited the maximum
absorption at 210 nm hence the wavelength was standardized in all studies. The main

282 JOURNAL OF COSMETIC SCIENCE
variations in the chromatographic conditions and lipids’ representative chromatogram of
the developed HPLC method are presented in Table II and Figure 3, respectively.
During the study, the major problem was peak overlapping, which is derived from the
physicochemical similarity of the phosphatidylcholine and Cer-NP.19 Namely, cholesterol
has saturated and unsaturated cyclic hydrocarbons. However, phosphatidylcholine and
Cer-NP similarly have a long fatty acid chain and polar head group, which state the
movement through the column.20 Furthermore, the steric hindrance derived from branching
and cycling indicates partitioning to the pore of column material hence, cholesterol might
have distinctness because of its cyclic hydrocarbons.12 Additionally, the functional groups
on the backbone affect the affinity of the analyte with the stationary phase.12 Herein, the
polarity of the head group on phosphatidylcholine could arise from its phosphate ions while
Cer-NP’s polarity was based on its hydroxyl groups.
On the other hand, the solubility of phosphatidylcholine, Cer-NP, and cholesterol have
differences based on their lipophilicities. Cer-NP has solubility problems in various solvents
and aqueous media.21 Even though cholesterol has a significantly lower lipophilic character
(log P of 8.7) rather than the other compounds (phosphatidylcholine: 12.9, Cer-NP: 12.4),
the chemical structure specified the retention time. Considering the chemical structure
similarity of phosphatidylcholine and Cer-NP, their retention times are expected to be very
similar. Hence, various compositions of mobile phase, flow rates, and injection volumes
were tested to prevent the overlapping of phosphatidylcholine, Cer-NP, and cholesterol
peaks, which could improve the retention times of the peaks of these compounds. The
retention times of lipid materials were studied in different conditions. In addition, tailing
factors (T
f )and the number of theoretical plates (N) of the Cer-NP peak were presented in
Table II. To improve peak resolution, the serial mobile phase composition (trials 1–5) was
tested with the mobile phase consisting of the mixture of methanol: acetonitrile from 60:40
(v/v) to 100:00 (v/v). These subsequent modifications in the mobile phase composition with
a reduction of its polarity caused an increase in lipid retention times. Trial 5 resulted in
low resolution (2.32) with the phosphatidylcholine peak still in conflict with the Cer-NP
peak. Thereby, the flow rate was fixed to be 0.8 mL/min in trials 12–14, and, then 0.5 mL/
min in trials 15–17. Also, the injection volume was decreased from 100 µL to 10 µL, to get
sharper peaks (trials: 12–17). The suitable method was repeated with a C8 column with
Figure 3. Representative HPLC chromatogram of the lipid components in skin-simulating liposome
formulation.